Department of Pharmacology & Toxicology

Acetaminophen (APAP) is a widely used analgesic, which is safe at therapeutic levels. APAP is mainly conjugated with glucuronic acid and sulfate to form water-soluble, nontoxic metabolites. Only a small portion of APAP is metabolized by P-450 isoenzymes, thereby forming the reactive metabolite N-acetyl-p-benzoquinone imine (NAPQI). NAPQI can react with sulfhydryl groups such as GSH. After APAP overdose, hepatic GSH is dramatically reduced, and NAPQI is able to bind to cellular proteins including mitochondrial proteins. This protein binding results in mitochondrial dysfunction and reactive oxygen species formation. This chain of events eventually leads to necrotic cell death of hepatocytes. My goal was to investigate the mechanisms and signaling pathways of APAP-induced cell necrosis in the liver, and to identify therapeutic approaches to prevent liver failure. Three protective strategies were investigated in detail: 1) Glutathione (GSH) and N-acetylcysteine (NAC) 2) Metallothionein (MT) 3) C-jun N-terminal kinase (JNK) inhibitor 1) Both in humans and in experimental animals, NAC is used as an antidote against APAP-induced liver injury. The doses of NAC that are being used clinically and experimentally are higher than needed for re-synthesis of hepatic GSH levels. In fact, our laboratory demonstrated that lower doses of GSH are highly effective in protecting against APAP toxicity. Therefore, I investigated whether there is a difference between the efficacy of NAC and GSH in protecting against APAP hepatotoxicity. Our data indicate that the amino acids supplied with the delayed treatment of the same dose of GSH or NAC are used for the re-synthesis of hepatic glutathione at similar levels, which protect against APAP-induced reactive oxygen species and peroxynitrite in the mitochondria. However, excess amino acids derived from GSH also serve as energy substrates for the Krebs cycle, which results in better protection against APAP hepatotoxicity than NAC treatment. Thus, the optimal protection by delayed GSH or NAC treatment involves the combination of two mechanisms, which are the accelerated recovery of mitochondrial GSH levels and the support of the mitochondrial bioenergetics. 2) Metallothionein (MT) expression attenuates APAP-induced liver injury; however, the mechanism of this protection remains incompletely understood. To address this issue, mice were treated with ZnCl2 for three days to induce MT. Twenty-four hours after the last dose of zinc, the animals received 300 mg/kg APAP. We found that the protective effect of MT in vivo was not due to the direct scavenging of reactive oxygen species and peroxynitrite. In addition, zinc treatment had no effect on the early GSH depletion kinetics after APAP administration, which is an indicator of the metabolic activation of APAP to its reactive metabolite NAPQI. MT was able to effectively trap NAPQI by covalent binding. We conclude that MT scavenges some of the excess NAPQI after GSH depletion and prevents covalent binding to cellular proteins,which is the trigger for the propagation of the cell injury mechanisms through mitochondrial dysfunction and nuclear DNA damage. 3) C-jun N-terminal kinase (JNK) has been suggested to contribute to APAP-induced liver injury. The postulated mechanism of JNK involvement was the promotion of mitochondrial Bax translocation, which triggers mitochondrial outer membrane pore formation and results in the release of intermembrane proteins such as apoptosis inducing factor (AIF) and endonuclease G (EndoG). However, it was reported that Bax-deficient mice were only temporally protected against APAP-induced liver injury (Bajt et al., 2008). In contrast, the protective effect of a JNK inhibitor was observed consistently up to 24 h. Therefore, additional mechanisms of injury involving JNK activation need to be considered. To address this issue, I treated mice with the JNK inhibitor, SP600125 1h before APAP (600 mg/kg). SP600125 reduced peroxynitrite formation; however, it did not have any significant effect on the level of nitrate and nitrite in plasma. Moreover, L-N-(1-iminoethyl)lysine (L-Nil), a specific iNOS inhibitor, attenuated neither plasma nitrate and nitrite levels nor hepatic injury after APAP injection. Taken together, SP600125 reduced peroxynitrite formation by decreasing superoxide formation. In summary, my investigation demonstrated that JNK is a critical factor for Bax translocation, which causes mitochondria outer membrane pore formation. In addition, JNK accelerates peroxynitrite generation via induction of superoxide formation. In conclusion, I demonstrated the efficacy of three protective strategies against APAP-induced hepatotoxicity: Mechanism of protection A: Preventing NAPQI binding to proteins (e.g., induction of MT gene expression); Mechanism of protection B: Scavenging (GSH, NAC) or reducing (JNK inhibition) the formation of reactive oxygen species; and Mechanism of protection C: Supplying mitochondrial energy substrates (GSH, NAC).

Many laboratory animal studies of migraine have employed electrophysiological techniques to assess neuronal sensitization, but few have examined behaviors using International Headache Society criteria, which are based on behavioral changes including duration and intensity of pain and avoidance of routine activity. Fewer still have attempted to correlate the appearance of these diagnostic symptoms with changes in the activity of pain-related neurotransmitters and neuromodulators, such as calcitonin gene related peptide (CGRP). A vasoactive neuropeptide, CGRP might contribute to the vasodilatory component of migraine, and to the pain associated with this condition as it is present in nociceptors, including those in the trigeminal ganglion that innervate cerebral vasculature. Previous work has shown that CGRP levels are increased in animal models of inflammatory pain and in the external jugular vein of humans during migraine attacks. The CGRP receptor is comprised of three proteins: a G-protein coupled receptor called calcitonin-like receptor (CLR), a receptor activity-modifying protein (RAMP1), and a coupling protein, receptor component protein (RCP). Thus, the availability and sensitivity of this receptor is subject to regulation at numerous levels. The objectives of this study were to develop a preclinical behavioral model of chronic migraine, to test sensory and motor behaviors relevant to International Headache Society diagnostic criteria, to examine whether there are sex differences in these behaviors, and to assess whether alterations in the expression pattern of genes encoding CGRP and its receptor components are associated with sex differences or changes in pain-related behaviors. Male and female Sprague-Dawley rats were implanted with a dural cannula placed over the occipital cortex. Groups of rats were treated with 10 or 20 microliter volumes of an inflammatory soup containing 1 mM each of histamine, serotonin, and bradykinin, as well as 0.1 mM prostaglandin E2 (pH 5.5). A control group received sterile phosphate-buffered saline (pH 7.4) alone. Baseline behavioral testing was conducted on all eight groups of animals prior to surgery and seven days later. The inflammatory soup or control solution was administered supradurally 3 times/week for a total of eight applications. Locomotor activity was assessed using force plate actimetry during and following application of the inflammatory soup or vehicle. Total RNA was isolated from ipsilateral trigeminal ganglia and ipsilateral medulla. Real-time polymerase chain reaction was used to quantify the expression of amplified constructs using gene specific primers for CGRP, RAMP1, CLR, and RCP. The results reveal pronounced sex differences in behavior following application of the inflammatory soup. Female rats displayed dose-dependent migraine-related behaviors and a longer duration of these effect in measurements of distance traveled, bouts of low mobility, and spatial confinement. Both males and females experienced allodynia following exposure to the inflammatory mixture. Moreover, females displayed a higher baseline gene expression of CGRP and lower baseline gene expression of RAMP1, CLR, and RCP in the medulla than male animals. In addition to these baseline differences, gene expression of CLR and RCP was induced in the medulla of female rats but not in males. No sex difference in CGRP gene expression was noted in the trigeminal ganglia, although females do have a lower baseline expression of CGRP receptors, RAMP1, CLR, and RCP than males. It was also found that in the trigeminal ganglia RAMP1, CLR and RCP are inducible, especially in males, and that at least a portion of these responses are the result of volume effects associated with application to the dura of the inflammatory soup or vehicle. These findings indicate significant sex dependent changes in rat locomotor activity and CGRP-related gene expression in the brainstem and trigeminal ganglia associated with the application to the dura of an inflammatory soup. As the behavioral endpoints utilized in this study correspond to clinical signs considered by the International Headache Society as diagnostic for migraine, these data confirm that CGRP and its receptors are involved in migraine pathophysiology and reveal for the first time that alterations in the response to this peptide may be related to the increased prevalence of migraine in females. Such findings could be of value in devising new therapeutic strategies for the treatment of the debilitating condition.

Many laboratory animal studies of migraine have employed electrophysiological techniques to assess neuronal sensitization, but few have examined behaviors using International Headache Society criteria, which are based on behavioral changes including duration and intensity of pain and avoidance of routine activity. Fewer still have attempted to correlate the appearance of these diagnostic symptoms with changes in the activity of pain-related neurotransmitters and neuromodulators, such as calcitonin gene related peptide (CGRP). A vasoactive neuropeptide, CGRP might contribute to the vasodilatory component of migraine, and to the pain associated with this condition as it is present in nociceptors, including those in the trigeminal ganglion that innervate cerebral vasculature. Previous work has shown that CGRP levels are increased in animal models of inflammatory pain and in the external jugular vein of humans during migraine attacks. The CGRP receptor is comprised of three proteins: a G-protein coupled receptor called calcitonin-like receptor (CLR), a receptor activity-modifying protein (RAMP1), and a coupling protein, receptor component protein (RCP). Thus, the availability and sensitivity of this receptor is subject to regulation at numerous levels. The objectives of this study were to develop a preclinical behavioral model of chronic migraine, to test sensory and motor behaviors relevant to International Headache Society diagnostic criteria, to examine whether there are sex differences in these behaviors, and to assess whether alterations in the expression pattern of genes encoding CGRP and its receptor components are associated with sex differences or changes in pain-related behaviors. Male and female Sprague-Dawley rats were implanted with a dural cannula placed over the occipital cortex. Groups of rats were treated with 10 or 20 microliter volumes of an inflammatory soup containing 1 mM each of histamine, serotonin, and bradykinin, as well as 0.1 mM prostaglandin E2 (pH 5.5). A control group received sterile phosphate-buffered saline (pH 7.4) alone. Baseline behavioral testing was conducted on all eight groups of animals prior to surgery and seven days later. The inflammatory soup or control solution was administered supradurally 3 times/week for a total of eight applications. Locomotor activity was assessed using force plate actimetry during and following application of the inflammatory soup or vehicle. Total RNA was isolated from ipsilateral trigeminal ganglia and ipsilateral medulla. Real-time polymerase chain reaction was used to quantify the expression of amplified constructs using gene specific primers for CGRP, RAMP1, CLR, and RCP. The results reveal pronounced sex differences in behavior following application of the inflammatory soup. Female rats displayed dose-dependent migraine-related behaviors and a longer duration of these effect in measurements of distance traveled, bouts of low mobility, and spatial confinement. Both males and females experienced allodynia following exposure to the inflammatory mixture. Moreover, females displayed a higher baseline gene expression of CGRP and lower baseline gene expression of RAMP1, CLR, and RCP in the medulla than male animals. In addition to these baseline differences, gene expression of CLR and RCP was induced in the medulla of female rats but not in males. No sex difference in CGRP gene expression was noted in the trigeminal ganglia, although females do have a lower baseline expression of CGRP receptors, RAMP1, CLR, and RCP than males. It was also found that in the trigeminal ganglia RAMP1, CLR and RCP are inducible, especially in males, and that at least a portion of these responses are the result of volume effects associated with application to the dura of the inflammatory soup or vehicle. These findings indicate significant sex dependent changes in rat locomotor activity and CGRP-related gene expression in the brainstem and trigeminal ganglia associated with the application to the dura of an inflammatory soup. As the behavioral endpoints utilized in this study correspond to clinical signs considered by the International Headache Society as diagnostic for migraine, these data confirm that CGRP and its receptors are involved in migraine pathophysiology and reveal for the first time that alterations in the response to this peptide may be related to the increased prevalence of migraine in females. Such findings could be of value in devising new therapeutic strategies for the treatment of the debilitating condition.

Acetaminophen (APAP) is a widely used analgesic, which is safe at therapeutic levels. APAP is mainly conjugated with glucuronic acid and sulfate to form water-soluble, nontoxic metabolites. Only a small portion of APAP is metabolized by P-450 isoenzymes, thereby forming the reactive metabolite N-acetyl-p-benzoquinone imine (NAPQI). NAPQI can react with sulfhydryl groups such as GSH. After APAP overdose, hepatic GSH is dramatically reduced, and NAPQI is able to bind to cellular proteins including mitochondrial proteins. This protein binding results in mitochondrial dysfunction and reactive oxygen species formation. This chain of events eventually leads to necrotic cell death of hepatocytes. My goal was to investigate the mechanisms and signaling pathways of APAP-induced cell necrosis in the liver, and to identify therapeutic approaches to prevent liver failure. Three protective strategies were investigated in detail: 1) Glutathione (GSH) and N-acetylcysteine (NAC) 2) Metallothionein (MT) 3) C-jun N-terminal kinase (JNK) inhibitor 1) Both in humans and in experimental animals, NAC is used as an antidote against APAP-induced liver injury. The doses of NAC that are being used clinically and experimentally are higher than needed for re-synthesis of hepatic GSH levels. In fact, our laboratory demonstrated that lower doses of GSH are highly effective in protecting against APAP toxicity. Therefore, I investigated whether there is a difference between the efficacy of NAC and GSH in protecting against APAP hepatotoxicity. Our data indicate that the amino acids supplied with the delayed treatment of the same dose of GSH or NAC are used for the re-synthesis of hepatic glutathione at similar levels, which protect against APAP-induced reactive oxygen species and peroxynitrite in the mitochondria. However, excess amino acids derived from GSH also serve as energy substrates for the Krebs cycle, which results in better protection against APAP hepatotoxicity than NAC treatment. Thus, the optimal protection by delayed GSH or NAC treatment involves the combination of two mechanisms, which are the accelerated recovery of mitochondrial GSH levels and the support of the mitochondrial bioenergetics. 2) Metallothionein (MT) expression attenuates APAP-induced liver injury; however, the mechanism of this protection remains incompletely understood. To address this issue, mice were treated with ZnCl2 for three days to induce MT. Twenty-four hours after the last dose of zinc, the animals received 300 mg/kg APAP. We found that the protective effect of MT in vivo was not due to the direct scavenging of reactive oxygen species and peroxynitrite. In addition, zinc treatment had no effect on the early GSH depletion kinetics after APAP administration, which is an indicator of the metabolic activation of APAP to its reactive metabolite NAPQI. MT was able to effectively trap NAPQI by covalent binding. We conclude that MT scavenges some of the excess NAPQI after GSH depletion and prevents covalent binding to cellular proteins,which is the trigger for the propagation of the cell injury mechanisms through mitochondrial dysfunction and nuclear DNA damage. 3) C-jun N-terminal kinase (JNK) has been suggested to contribute to APAP-induced liver injury. The postulated mechanism of JNK involvement was the promotion of mitochondrial Bax translocation, which triggers mitochondrial outer membrane pore formation and results in the release of intermembrane proteins such as apoptosis inducing factor (AIF) and endonuclease G (EndoG). However, it was reported that Bax-deficient mice were only temporally protected against APAP-induced liver injury (Bajt et al., 2008). In contrast, the protective effect of a JNK inhibitor was observed consistently up to 24 h. Therefore, additional mechanisms of injury involving JNK activation need to be considered. To address this issue, I treated mice with the JNK inhibitor, SP600125 1h before APAP (600 mg/kg). SP600125 reduced peroxynitrite formation; however, it did not have any significant effect on the level of nitrate and nitrite in plasma. Moreover, L-N-(1-iminoethyl)lysine (L-Nil), a specific iNOS inhibitor, attenuated neither plasma nitrate and nitrite levels nor hepatic injury after APAP injection. Taken together, SP600125 reduced peroxynitrite formation by decreasing superoxide formation. In summary, my investigation demonstrated that JNK is a critical factor for Bax translocation, which causes mitochondria outer membrane pore formation. In addition, JNK accelerates peroxynitrite generation via induction of superoxide formation. In conclusion, I demonstrated the efficacy of three protective strategies against APAP-induced hepatotoxicity: Mechanism of protection A: Preventing NAPQI binding to proteins (e.g., induction of MT gene expression); Mechanism of protection B: Scavenging (GSH, NAC) or reducing (JNK inhibition) the formation of reactive oxygen species; and Mechanism of protection C: Supplying mitochondrial energy substrates (GSH, NAC).

Polybrominated diphenyl ethers (PBDEs) were introduced in the late 1970's as additive flame retardants incorporated into textiles, electronics, plastics and furniture. Although 2,2',3,3',4,4',5,5',6,6'-decabromodiphenyl ether (BDE209) is the only congener currently on the market, 2,2`,4,4`-tetrabromodiphenyl ether (BDE47), 2,2`,4,4`,5-pentabromodiphenyl ether (BDE99), and 2,2`,4,4`,5,5`-hexabromodiphenyl ether (BDE153) are the predominant congeners detected in human and wildlife samples. Upon exposure, PBDEs enter the liver where they are biotransformed to potentially toxic metabolites. Although the human liver burden of PBDEs is not clear, the presence of PBDEs in human liver is particularly alarming because it has been demonstrated in rodents that hydroxylated metabolites may play a pivotal role in PBDE-mediated toxicity. The mechanism by which PBDEs enter the liver was not known. However, due to their large molecular weights (MWs ~485 to 1000 Da), they were not likely to enter hepatocytes by simple diffusion. Organic anion transporting polypeptides (OATPs: human; Oatps: rodents) are responsible for hepatic uptake of a variety of amphipathic compounds of MWs larger than 350 Da. Therefore, I tested the hypothesis that OATPs/Oatps expressed in human and mouse hepatocytes are responsible for the uptake of PBDE congeners 47, 99, and 153 by using Chinese hamster ovary (CHO) cell lines expressing OATP1B1, OATP1B3, or OATP2B1 and Human Embryonic Kidney 293 (HEK293) cells transiently expressing Oatp1a1, Oatp1a4, Oatp1b2, or Oatp2b1. Direct uptake studies illustrated that PBDE congeners are substrates of human and mouse hepatic OATPs/Oatps, except for Oatp1a1. Detailed kinetic analysis revealed that OATP1B1, OATP1B3, Oatp1a4, and Oatp1b2 transport BDE47 with the highest affinity followed by BDE99 and BDE153. However, both OATP2B1 and Oatp2b1 transported all three congeners with similar affinities. The importance of hepatic Oatps for the accumulation of BDE47 in liver was confirmed using Oatp1a4- and Oatp1b2-null mice. These results clearly suggest that uptake of PBDEs via these OATPs/Oatps are responsible for liver-specific accumulation of PBDEs. In mouse liver, PBDEs induce drug metabolizing enzymes, namely cytochrome P450s (Cyps). However, the molecular mechanisms underlying this induction was unknown. Cyp2b10 and 3a11 are target genes of the xenobiotic nuclear receptors, the constitutive androstane receptor (CAR) and pregnane X receptor (PXR), both of which are responsible for mediating induction of Cyp2b10 and Cyp3a11, respectively. I hypothesized that PBDE congeners are CAR and/or PXR activators. Using reporter-gene luciferase assays I showed that BDE47, BDE99 and BDE209 activate human and mouse CAR and PXR in a concentration-dependent manner. Furthermore, induction of Cyp2b10 and Cyp3a11 was markedly suppressed in CAR- and PXR-null mice, respectively, indicating that PBDE congeners activate these receptors in vivo. BDE47 and BDE99, the primary congeners detected in humans in the United States, are capable of inducing Cyp2b and Cyp3a enzymes in rodents. However, it is not clear which Cyp isoform, if any, is preferentially induced upon exposure to BDE47 or BDE99. Induction of mouse hepatic Cyp2b10 and 3a11 by PBDEs showed distinct dose-responses, with Cyp2b10 being induced at lower doses and Cyp3a11 at much higher doses, indicating PBDEs are more likely to induce hepatic enzymes at doses that humans are exposured to. Currently, daily exposure of PBDEs is estimated to be 0.003mg/kg for adults. This study shows that effects of PBDEs are seen in animal models at concentrations within ~10-fold of the high end of the human population. Together, the results from the current dissertational study demonstrate that PBDEs are substrates of OATPs/Oatps and activators of CAR and PXR. This study not only provides a molecular basis for understanding PBDE disposition and toxicity in the liver but also cautions PBDE exposure may result in broader impact on liver physiology and toxicology.

The organic anion transporting polypeptides (human: OATP; other: Oatp) form a mammalian transporter superfamily that mediates the transport of structurally unrelated compounds across the cell membrane. Members in this superfamily participate in the absorption, distribution and excretion of many endogenous and exogenous substances including a number of medications and environmental toxicants. Polymorphisms of OATPs have been shown clinically to give rise to inter-individual variabilities of drug efficacy and/or toxicity. Furthermore, as multi-specific transporters, they are potential sites for drug-drug interactions. Therefore, understanding the mechanism of OATP/Oatp mediated transport of endo- and xenobiotics will not only help to improve drug efficacy but also to improve the prediction and prevention of toxicity. The overall goal of this dissertation is identifying key amino acids that may play an important role in OATP/Oatp-mediated transport and investigating the spatial size of the substrate binding/translocation pocket. In this dissertation, I defended three specific aims. In the first specific aim, I evaluated the hypothesis that conserved positively charged amino acids play important roles in OATP1B1 transport function. To address this aim, site-directed mutagenesis was employed and the mutants of several conserved positively charged amino acids were studied. The two extracellular amino acids R57 and K361 were found to be important in OATP1B1 mediated transport of estradiol-17β-glucuronide, estrone-3-sulfate and BSP. In the second specific aim, I evaluated the hypothesis that quantifying transport activities of different substrates mediated by chimeras between rat Oatp1a1 and Oatp1a4 in combination with site-directed mutagenesis should allow us to identify regions and/or individual amino acids that are important for Oatp1a4-mediated substrate recognition and/or transport. The effects of chimeric proteins on transport activity were substrate dependent. Extracellular loop 4 and transmembrane domain 8 were identified to be important in transport of digoxin, taurocholate and estradiol-17β-glucuronide. The C-terminal half of Oatp1a4 and Oatp1a1 was found to be important for BSP transport and the interactions between the N-terminal half and the C-terminal half of Oatp1a4 is essential for DPDPE transport. In the third specific aim, I evaluated the hypothesis that different rat renal organic anion transporters of the Oat and Oatp families selectively transport perfluorinated carboxylates (PFCAs) depending on the chain lengths. The purpose of this study was to determine the substrate size selectivity of Oats and Oatp1a1. To address this aim, the inhibitory effects of PFCAs with chain length from C2 to C18 on transport of model substrates by rat Oat1, Oat2, Oat3, Urat1 and Oatp1a1 was quantified. Furthermore, direct uptake of the best inhibitors was characterized. The best substrates for Oats were C7 and C8, whereas Oatp1a1 transported longer PFCAs such as C9 and C10 better than C8 and C7. Altogether, this dissertation reveals that (1) some conserved positively charged amino acids and the C-terminal half are important for OATP/Oatp mediated transport of certain substrates; (2) OATP/Oatps have several binding/translocation sites for different substrates and (3) the preferred substrate size for OATP/Oatps is slightly bigger than that for Oats.

Topically applied morphine is routinely used to alleviate pain in cutaneous wounds such as burns and pressure sores, yet evidence suggests the topical administration of exogenous opioid drugs may impair wound closure. This dissertation research was designed to test the hypothesis that topical morphine application delays cutaneous wound healing via mechanisms dependent upon peripheral neuropeptide activity. Results demonstrate that topical morphine application delays cutaneous wound closure rates. The delay occurs in a concentration-dependent manner (consistent with opioid-receptor mediated effects), is mimicked by NK-1 and NK-2 receptor antagonists, and can be reversed by the exogenous application of either substance P or neurokinin A. The results indicate that morphine acts presynaptically, delaying wound closure by activating opioid receptors located on primary afferent nerve terminals and subsequently inhibiting the antidromic release of neuropeptides into the wound. The temporal pattern of the effects of topical morphine treatment can be attributed to alterations in the initiation and duration of essential, early processes during wound healing. The delay in closure evoked by topical morphine not only leaves the cutaneous wound open longer, increasing the risk of infection, but also results in long-term architectural deficits, compromising the integrity of the healed skin. Furthermore, dysregulation of neurokinin receptor-expressing cells essential for normal wound healing emerges as a mechanism capable of significantly disrupting the dynamic processes involved in wound healing.